16 bit v. 8 bit

len20

I recently purchased CS4 for the first time and I'm learning about 8-bit v. 16-bit channels. My simple understanding is that 16-bit is better because it will provide more information and therefore more accurate color. However, jpeg format does not allow 16-bit, Smugmug does not allow prints to be made from anything but jpeg (eg. tiff files), and therefore my prints will always be made from 8-bit rather than 16-bit files.

My question: should I be bothered by this? Does it effect the quality of the prints?

I apologize if this issue has been covered elsewhere. I did my best to search in the forums.

Andy

len20 wrote:

My question: should I be bothered by this? Does it effect the quality of the prints?

No and if you produce quality files, good exposures, color, and good processing, it 8 bit won't affect your prints, either

jfriend

If your shoot a proper exposure and don't need any significant post processing moves to correct your image, then I'd agree with Andy and say that you won't find any difference between 8-bit and 16-bit processing.

However, the place where it can make a meaningful difference is when everything did not go right in the shot and you do need to make some significant corrections. In those cases, capturing more than 8-bits of data (by shooting RAW) and then using 16-bits to manipulate that data can give you a better result.

Once you've obtained your final result, there is little sacrifice in converting that to 8-bits for printing or web viewing where the output devices are likely all 8-bits anyway.

There are also certain exotic post processing moves that it is recommended to do in 16-bits, even if the original wasn't in 16-bits (such as certain LAB colorspace adjustments), but that's a more advanced topic.

Art Scott

jfriend wrote:

If your shoot a proper exposure and don't need any significant post processing moves to correct your image, then I'd agree with Andy and say that you won't find any difference between 8-bit and 16-bit processing.

However, the place where it can make a meaningful difference is when everything did not go right in the shot and you do need to make some significant corrections. In those cases, capturing more than 8-bits of data (by shooting RAW) and then using 16-bits to manipulate that data can give you a better result.

Once you've obtained your final result, there is little sacrifice in converting that to 8-bits for printing or web viewing where the output devices are likely all 8-bits anyway.

There are also certain exotic post processing moves that it is recommended to do in 16-bits, even if the original wasn't in 16-bits (such as certain LAB colorspace adjustments), but that's a more advanced topic.

I agreethumbbow

Or as in my case I had a publisher wanting files for printing asked for them as 16 bit tiffs .....I could not supply as everything I had was in 8 bit......that was one of those never again moments I have had since going digital......I also keep everything in a colorspace like Adobe RGB rahter than sRGB for the same reasons.......

len20

Thank you. This was helpful. It sounds like as long as I've got the RAW file I can go back and process the image in using 16-bit channel if I needed to for some reason. Otherwise, I'll just output them in 8-bit jpegs and not have to worry about a loss in quality. Just what I needed to hear.